The emergence of new vehicles, such as optionally piloted vehicles (OPVs) or unmanned aerial vehicles (UAVs), require new techniques for modeling a vehicle's environment. For example, sensors are used to capture data related to the external environment of the vehicle. These sensors may vary in terms of, e.g., the type of data they collect and the amount of data they collect.
Efficiency is beneficial in terms of representation, classification, and storage of sensor data. Current schemes for organizing the data are only useful for a particular sensor or for a very specific task. For example, to be able to execute even a simple resupply mission completely autonomously, an environmental model must be able to represent and classify a wide variety of inputs and data types.